Gleich, GJ

Abstract The syndrome of episodic angioedema and eosinophilia is characterized by cyclic edema, marked peripheral blood eosinophilia, and eosinophil degranulation in the dermis. Using a sensitive immunoenzymetric method, we measured serum interleukin (IL)-5 levels in four patients with this syndrome. We also determined the percentage of activated T cells in the peripheral blood of a new patient before and during an attack. In the patient presented, IL-5 levels peaked several days before maximal eosinophilia and then declined. This patient's lymphocytes showed an increased percentage, 28% (normal 2% to 3%), of activated T cells staining for both CD3 and HLA-DR 10 days before maximal eosinophilia, but no increase at the time of peak eosinophilia. In serum from three previously reported cases, elevated serum IL-5 levels were found during attacks. After glucocorticoid administration, IL-5 levels became undetectable in three of the four patients. Production of IL-5 is likely an important determinant of the pathophysiology of this syndrome.

Abstract We investigated the ultrastructural characteristics and the granule major basic protein (MBP) content of hypodense eosinophils from patients with the hypereosinophilic syndrome who had at least 90% hypodense eosinophils in their peripheral blood and compared these cells to normodense eosinophils from normal persons. The hypodense cells (density less than 1.082) contained significantly less MBP than normodense (density greater than 1.082) eosinophils (P less than .001) as measured by radioimmunoassay (RIA). Electron microscopic examination demonstrated a mean of 25.0 +/- 4.4 (X +/- 1 SD) granules per hypodense cell, compared to 30.6 +/- 8.4 granules per cell in the normodense group (P less than .1). The most striking difference between the hypodense and normodense eosinophils was the small individual granule size (X = .14 +/- .05 v .26 +/- .05 micron 2, respectively, P less than .001), and the smaller total granule area (3.2 +/- 1.8 vs 7.7 +/- 3.1 micron 2, respectively, P less than .001). Because the cytoplasmic areas were similar in the two groups, the mean percent area of cytoplasm occupied by granules was significantly lower in the hypodense group (P less than .001). The finding of consistently smaller granules in the presence of equal or fewer granules per cell in the hypodense eosinophils may explain the lower MBP content and thus provide a morphologic basis for the low density of eosinophils in patients with the hypereosinophilic syndrome.

Abstract Peripheral eosinophilia is almost invariably observed during the course of interleukin-2 (IL-2) therapy and is frequently accompanied by the development of a capillary leak syndrome characterized by edema, weight gain, and oliguria. We studied five patients with advanced malignancy treated with IL-2. Eosinophilia was not present initially but developed in all patients late in the course of therapy, with counts ranging from 2,328/mm3 to 15,958/mm3. In all patients, there was a temporal relationship between the infusion of IL-2 and the appearance of elevated plasma concentrations of IL-5, a growth factor for eosinophils. Granulocyte-macrophage colony-stimulating factor was not detectable in plasma. IL-4 and gamma-interferon plasma levels were variably elevated. Plasma concentrations of major basic protein, a toxic eosinophil granule protein, began increasing before eosinophil counts increased. By the time of the third IL-2 infusion, high concentrations of major basic protein were present in all five patients (up to 5,600 ng/mL) and skin biopsies showed major basic protein deposition in the dermis. Four patients developed significant capillary leak syndrome and all of these patients showed markedly elevated major basic protein levels. The lowest peak plasma concentration of major basic protein (1,751 ng/mL) was observed in the one patient who did not develop edema and weight gain. These results suggest that IL-2 induces IL-5 leading to marked peripheral eosinophilia and extravascular eosinophil degranulation. The release of toxic eosinophil products at extravascular sites and in the circulation may contribute to the pathogenesis of the capillary leak syndrome complicating IL-2 therapy.

The syndrome of episodic angioedema and eosinophilia is characterized by cyclic edema, marked peripheral blood eosinophilia, and eosinophil degranulation in the dermis. Using a sensitive immunoenzymetric method, we measured serum interleukin (IL)-5 levels in four patients with this syndrome. We also determined the percentage of activated T cells in the peripheral blood of a new patient before and during an attack. In the patient presented, IL-5 levels peaked several days before maximal eosinophilia and then declined. This patient's lymphocytes showed an increased percentage, 28% (normal 2% to 3%), of activated T cells staining for both CD3 and HLA-DR 10 days before maximal eosinophilia, but no increase at the time of peak eosinophilia. In serum from three previously reported cases, elevated serum IL-5 levels were found during attacks. After glucocorticoid administration, IL-5 levels became undetectable in three of the four patients. Production of IL-5 is likely an important determinant of the pathophysiology of this syndrome.

Peripheral eosinophilia is almost invariably observed during the course of interleukin-2 (IL-2) therapy and is frequently accompanied by the development of a capillary leak syndrome characterized by edema, weight gain, and oliguria. We studied five patients with advanced malignancy treated with IL-2. Eosinophilia was not present initially but developed in all patients late in the course of therapy, with counts ranging from 2,328/mm3 to 15,958/mm3. In all patients, there was a temporal relationship between the infusion of IL-2 and the appearance of elevated plasma concentrations of IL-5, a growth factor for eosinophils. Granulocyte-macrophage colony-stimulating factor was not detectable in plasma. IL-4 and gamma-interferon plasma levels were variably elevated. Plasma concentrations of major basic protein, a toxic eosinophil granule protein, began increasing before eosinophil counts increased. By the time of the third IL-2 infusion, high concentrations of major basic protein were present in all five patients (up to 5,600 ng/mL) and skin biopsies showed major basic protein deposition in the dermis. Four patients developed significant capillary leak syndrome and all of these patients showed markedly elevated major basic protein levels. The lowest peak plasma concentration of major basic protein (1,751 ng/mL) was observed in the one patient who did not develop edema and weight gain. These results suggest that IL-2 induces IL-5 leading to marked peripheral eosinophilia and extravascular eosinophil degranulation. The release of toxic eosinophil products at extravascular sites and in the circulation may contribute to the pathogenesis of the capillary leak syndrome complicating IL-2 therapy.

We investigated the ultrastructural characteristics and the granule major basic protein (MBP) content of hypodense eosinophils from patients with the hypereosinophilic syndrome who had at least 90% hypodense eosinophils in their peripheral blood and compared these cells to normodense eosinophils from normal persons. The hypodense cells (density less than 1.082) contained significantly less MBP than normodense (density greater than 1.082) eosinophils (P less than .001) as measured by radioimmunoassay (RIA). Electron microscopic examination demonstrated a mean of 25.0 +/- 4.4 (X +/- 1 SD) granules per hypodense cell, compared to 30.6 +/- 8.4 granules per cell in the normodense group (P less than .1). The most striking difference between the hypodense and normodense eosinophils was the small individual granule size (X = .14 +/- .05 v .26 +/- .05 micron 2, respectively, P less than .001), and the smaller total granule area (3.2 +/- 1.8 vs 7.7 +/- 3.1 micron 2, respectively, P less than .001). Because the cytoplasmic areas were similar in the two groups, the mean percent area of cytoplasm occupied by granules was significantly lower in the hypodense group (P less than .001). The finding of consistently smaller granules in the presence of equal or fewer granules per cell in the hypodense eosinophils may explain the lower MBP content and thus provide a morphologic basis for the low density of eosinophils in patients with the hypereosinophilic syndrome.